26 Apr, 2007
32 commits
-
Writing to /sys/class/net/brX/bridge/stp_state causes a warning because
RTNL is not held when call br_stp_if.cSigned-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
If a bridge is not running STP, then it has no way to detect a cycle
in the network. But if it is not running STP and some other machine
or device is running STP, then if STP BPDU's get forwarded to it can
detect the cycle.This is how the old 2.4 and early 2.6 code worked.
Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
Pause frames should never make it out of the network device into
the stack. But if a device was misconfigured, it might happen.
So drop pause frames in bridge.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
The change to forward STP bpdu's (for usermode STP) through normal path,
changed the packet type in the process. Since link local stuff is multicast, it
should stay pkt_type = PACKET_MULTICAST. The code was probably copy/pasted
incorrectly from the bridge pseudo-device receive path.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
The following is leftover from earlier change in net-2.6.22.
Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
Since we're now holding the rtnl during the entire dump operation, we can
remove additional locking for rtnl protected data. This patch does that
for all simple cases (dev_base_lock for dev_base walking, RCU protection
for FIB rule dumping).Signed-off-by: Patrick McHardy
Signed-off-by: David S. Miller -
Switch cb_lock to mutex and allow netlink kernel users to override it
with a subsystem specific mutex for consistent locking in dump callbacks.
All netlink_dump_start users have been audited not to rely on any
side-effects of the previously used spinlock.Signed-off-by: Patrick McHardy
Signed-off-by: David S. Miller -
All LOG targets always use their internal logging function nowadays, so
remove the incorrect error message and handle real errors (!= -EEXIST)
by failing to load.Signed-off-by: Patrick McHardy
Signed-off-by: David S. Miller -
The attached patch adds gratuitous arp filtering, more precisely: it
allows checking that the IPv4 source address matches the IPv4
destination address inside the ARP header. It also adds a check for the
hardware address type when matching MAC addresses (nothing critical,
just for better consistency).Signed-off-by: Bart De Schuymer
Acked-by: Carl-Daniel Hailfinger
Signed-off-by: Patrick McHardy
Signed-off-by: David S. Miller -
The attached patch by Michael Milner adds support for using iptables and
ip6tables on bridged traffic encapsulated in ppoe frames, similar to
what's already supported for vlan.Signed-off-by: Michael Milner
Signed-off-by: Bart De Schuymer
Signed-off-by: Patrick McHardy
Signed-off-by: David S. Miller -
This patch checks kmem_cache_create() error and aborts loading module
on failure.Signed-off-by: Akinobu Mita
Signed-off-by: Stephen Hemminger -
For case of bridging pseudo devices, the get created/destroyed (Xen)
need to allow setting address to any valid value.Signed-off-by: Stephen Hemminger
-
Need to tell STP daemon about more events, like any time a
device is added even when it is down.Signed-off-by: Stephen Hemminger
-
This patchset based on work by Aji_Srinivas@emc.com provides allows
spanning tree to be controled from userspace. Like hotplug, it
uses call_usermodehelper when spanning tree is enabled so there
is no visible API change. If call to start usermode STP fails
it falls back to existing kernel STP.Signed-off-by: Stephen Hemminger
-
The RSTP daemon needs to be able to flush all dynamic forwarding
entries in the case of topology change.This is a temporary interface. It will change to a netlink interface
before RSTP daemon is officially released.Signed-off-by: Stephen Hemminger
-
Instead of hashing the whole Ethernet address, it should be faster
to just use the last 4 bytes. Add a random salt value to the hash
to make it more difficult to construct worst case DoS hash chains.Signed-off-by: Stephen Hemminger
-
While in the STP learning state, don't route packets; wait until
forwarding delay has expired. The purpose of the forwarding delay
is to detect loops in the network, and if a brouter started up
and started forwarding, it could cause a flood.Signed-off-by: Stephen Hemminger
-
Change the bridging hook to be simple function with return value
rather than modifying the skb argument. This could generate better
code and is cleaner.Signed-off-by: Stephen Hemminger
-
To clearly state the intent of copying to linear sk_buffs, _offset being a
overly long variant but interesting for the sake of saving some bytes.Signed-off-by: Arnaldo Carvalho de Melo
-
To clearly state the intent of copying from linear sk_buffs, _offset being a
overly long variant but interesting for the sake of saving some bytes.Signed-off-by: Arnaldo Carvalho de Melo
-
Right now Xen has a horrible hack that lets it forward packets with
partial checksums. One of the reasons that CHECKSUM_PARTIAL and
CHECKSUM_COMPLETE were added is so that we can get rid of this hack
(where it creates two extra bits in the skbuff to essentially mirror
ip_summed without being destroyed by the forwarding code).I had forgotten that I've already gone through all the deivce drivers
last time around to make sure that they're looking at ip_summed ==
CHECKSUM_PARTIAL rather than ip_summed != 0 on transmit. In any case,
I've now done that again so it should definitely be safe.Unfortunately nobody has yet added any code to update CHECKSUM_COMPLETE
values on forward so we I'm setting that to CHECKSUM_NONE. This should
be safe to remove for bridging but I'd like to check that code path
first.So here is the patch that lets us get rid of the hack by preserving
ip_summed (mostly) on forwarded packets.Signed-off-by: Herbert Xu
Signed-off-by: David S. Miller -
Signed-off-by: Patrick McHardy
Signed-off-by: David S. Miller -
Signed-off-by: Thomas Graf
Signed-off-by: David S. Miller -
Renaming skb->h to skb->transport_header, skb->nh to skb->network_header and
skb->mac to skb->mac_header, to match the names of the associated helpers
(skb[_[re]set]_{transport,network,mac}_header).Signed-off-by: Arnaldo Carvalho de Melo
Signed-off-by: David S. Miller -
Now the skb->nh union has just one member, .raw, i.e. it is just like the
skb->mac union, strange, no? I'm just leaving it like that till the transport
layer is done with, when we'll rename skb->mac.raw to skb->mac_header (or
->mac_header_offset?), ditto for ->{h,nh}.Signed-off-by: Arnaldo Carvalho de Melo
Signed-off-by: David S. Miller -
Signed-off-by: Arnaldo Carvalho de Melo
Signed-off-by: David S. Miller -
Use logic operations rather than memcmp() to compare destination
address with link local multicast addresses.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
Signed-off-by: Arnaldo Carvalho de Melo
Signed-off-by: David S. Miller -
For the places where we need a pointer to the network header, it is still legal
to touch skb->nh.raw directly if just adding to, subtracting from or setting it
to another layer header.Signed-off-by: Arnaldo Carvalho de Melo
Signed-off-by: David S. Miller -
For the places where we need a pointer to the mac header, it is still legal to
touch skb->mac.raw directly if just adding to, subtracting from or setting it
to another layer header.This one also converts some more cases to skb_reset_mac_header() that my
regex missed as it had no spaces before nor after '=', ugh.Signed-off-by: Arnaldo Carvalho de Melo
Signed-off-by: David S. Miller -
For the common, open coded 'skb->mac.raw = skb->data' operation, so that we can
later turn skb->mac.raw into a offset, reducing the size of struct sk_buff in
64bit land while possibly keeping it as a pointer on 32bit.This one touches just the most simple case, next will handle the slightly more
"complex" cases.Signed-off-by: Arnaldo Carvalho de Melo
Signed-off-by: David S. Miller -
We currently use a special structure (struct skb_timeval) and plain
'struct timeval' to store packet timestamps in sk_buffs and struct
sock.This has some drawbacks :
- Fixed resolution of micro second.
- Waste of space on 64bit platforms where sizeof(struct timeval)=16I suggest using ktime_t that is a nice abstraction of high resolution
time services, currently capable of nanosecond resolution.As sizeof(ktime_t) is 8 bytes, using ktime_t in 'struct sock' permits
a 8 byte shrink of this structure on 64bit architectures. Some other
structures also benefit from this size reduction (struct ipq in
ipv4/ip_fragment.c, struct frag_queue in ipv6/reassembly.c, ...)Once this ktime infrastructure adopted, we can more easily provide
nanosecond resolution on top of it. (ioctl SIOCGSTAMPNS and/or
SO_TIMESTAMPNS/SCM_TIMESTAMPNS)Note : this patch includes a bug correction in
compat_sock_get_timestamp() where a "err = 0;" was missing (so this
syscall returned -ENOENT instead of 0)Signed-off-by: Eric Dumazet
CC: Stephen Hemminger
CC: John find
Signed-off-by: David S. Miller
18 Apr, 2007
1 commit
-
compare_ether_addr() implicitly requires that the addresses
passed are 2-bytes aligned in memory.This is not true for br_stp_change_bridge_id() and
br_stp_recalculate_bridge_id() in which one of the addresses
is unsigned char *, and thus may not be 2-bytes aligned.Signed-off-by: Evgeny Kravtsunov
Signed-off-by: Kirill Korotaev
Signed-off-by: Pavel Emelianov
23 Mar, 2007
2 commits
-
Turning up the warnings on gcc makes it emit warnings
about the placement of 'inline' in function declarations.
Here's everything that was under net/Signed-off-by: Dave Jones
Signed-off-by: David S. Miller -
br_fdb_get use atomic_inc to increase the refcount of an element found
on a RCU protected list, which can lead to the following race:CPU0 CPU1
br_fdb_get: rcu_read_lock
__br_fdb_get: find element
fdb_delete: hlist_del_rcu
br_fdb_put
br_fdb_put: atomic_dec_and_test
call_rcu(fdb_rcu_free) br_fdb_get: atomic_inc
rcu_read_unlock
fdb_rcu_free: kmem_cache_freeUse atomic_inc_not_zero instead.
Signed-off-by: Patrick McHardy
Signed-off-by: David S. Miller
08 Mar, 2007
1 commit
-
One change introduced by the workqueue removal patch is that adding an
interface that is up to a bridge which is also up does not ever call
br_stp_enable_port(), leaving the port in DISABLED state until we do
ifconfig down and up or link events occur.The following patch to the br_add_if function fixes it.
This is a regression introduced in 2.6.21.Submitted-by: Aji_Srinivas@emc.com
Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
01 Mar, 2007
1 commit
-
This change goes with earlier change to get rid of
work queue for path cost. Now stp_set_path_cost does its own
locking. This is to allow it to call br_path_cost() which calls
ethtool interfaces (might sleep).Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
27 Feb, 2007
2 commits
-
Having a work queue for checking carrier leads to lots of race issues.
Simpler to just get the cost when data structure is created and
update on change.Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller -
The bridge hasn't used miscdevice for a long long time.
Signed-off-by: Stephen Hemminger
Signed-off-by: David S. Miller
21 Feb, 2007
1 commit
-
We frequently need the maximum number of possible processors in order to
allocate arrays for all processors. So far this was done using
highest_possible_processor_id(). However, we do need the number of
processors not the highest id. Moreover the number was so far dynamically
calculated on each invokation. The number of possible processors does not
change when the system is running. We can therefore calculate that number
once.Signed-off-by: Christoph Lameter
Cc: Frederik Deweerdt
Cc: Neil Brown
Cc: Trond Myklebust
Signed-off-by: Andrew Morton
Signed-off-by: Linus Torvalds
